The long-term goal of this project is to obtain a better understanding of the molecular mechanisms involved in cellular immune phenomena. Our present efforts are concentrated on the cytotoxicity of human natural killer (NK) lymphocytes for cultured tumor cell lines. We previously developed a kinetic method and distribution-free statistical procedure that allow for the precise determination of the maximal velocity and apparent Michaelis constant for natural cytotoxicity. Recently, we also developed procedures that allow for the measurement of the rate constants for the individual steps of the cytotoxicity reaction, i.e., binding, lytic programming, and killer cell-independent lysis (KCIL). Values obtained for these parameters in parallel assays using identical donor and target cells have been used to test the experimental validity of a multistep kinetic model for natural killing. Briefly, values for the rate of lytic programming, Vmax, and the frequency of target-binding cells (obtained in single-cell cytotoxicity assays) can be used to determine the frequency of active NK cells in the effector cell population. This value, coupled with our finding that the rate of KCIL is approximately 10 times faster than the rate of lytic programming for K-562 target cells, has allowed us to determine that (for K-562 targets) the apparent Michaelis constant is approximately equal to the frequency of target binding cells. Furthermore, using these methods, we generated strong evidence that the primary effect of interferon on NK cells is to increase their rate of lytic programming. We have also used a kinetic approach to investigate the specificity of NK cytotoxicity with clear evidence for purely competitive inhibition by heterologous NK target cells. This indicates multiple target antigens, some of which are shared by different NK targat cells. Finally, we developed a nonisotopic cytotoxicity assay in which the release of lactate dehydrogenase by lysed K-562 cells is monitored and find excellent agreement between this method and the more expensive and time-consuming Cr-release assay. (LB)